Aero-thermodynamic responses of a novel FGM sector disks using mathematical modeling and deep neural networks.

• Presenting aero-thermodynamic responses MD-FG sector disk in supersonic airflow, for the first time. • Presenting quasi-3D new refined theory to present the displacement fields of the MD-FG sector disk in higher-order shapes. • Presenting flutter and frequency response of the temperature-dependent MD-FG sector disk for different parameters. • Presenting the GDQ method to solve the governing equations of the current higher-order structure. • Presenting deep neural networks model with related parameters to simulate the current applicable structure with low computational cost. Composite sector disks have extensive applications in aerospace industries, particularly when exposed to challenging conditions such as supersonic airflow and thermal environments. These applications leverage the superior properties of composite materials, including high strength-to-weight ratios, enhanced durability, and excellent thermal resistance, to meet the stringent requirements of aerospace operations. Multi-directional functionally graded (MD-FG) materials due to high-temperature resistance and other amazing properties in each direction have gotten plenty of attention recently. So, in this research, a thermoelasticity solution has been presented to study fundamental frequency traits of an MD-FG sector disk in supersonic airflow via both mathematics simulation and deep neural networks technique. For obtaining exact displacement fields, along with defining the changes of transverse shear strains along the system's thickness, the refined zigzag hypothesis is utilized. For obtaining the temperature-dependent equations, heat conduction relation and thermal boundary conditions of the MD-FG structure are presented. A coupled quasi-3D new refined theory (Q3D-NRT) and generalized differential quadrature method (GDQM) are presented for obtaining and solving the partial differential equations in the time-displacement domain. After obtaining the mathematics results, appropriate datasets are made for testing, training, and validation of the deep neural networks technique. Finally, the results have shown that aerodynamic pressure, temperature changes, Mach number, free stream speed, and air yaw angle have a major role in the stability/instability analyses of the thermally affected MD-FG sector disk in supersonic airflow. As an amazing outcome, increasing the sector angle, FG indexes, and temperature change lead to the reduction of the critical Mach number, and aerodynamic pressure associated with the flutter phenomenon. [ABSTRACT FROM AUTHOR]